Optical communication systems have become of critical importance within today's society. With this increase in importance comes a need for the optical communication infrastructures to maintain a high reliability and signal quality. To create these high standards, many techniques have been implemented. One extremely successful and wide spread method to maintain the integrity of communication networks, even during problem time periods, is through the use of line protection.
The key principle behind line protection within optical communication networks is the generation of working and protection channels for each communication link, the working and protection channels being physically identical in both bandwidth and functionality. In normal operations, with no faults indicated on either channel, the communication path is chosen to be through the working channel. The use of line protection allows communications that are traversing a working channel to be transferred to a protection channel when problems occur with respect to the working channel, with minimal interruption in the actual transmission of information. There are numerous different configurations for line protection systems that generally create a trade-off between the number of optical fibre cables (hereinafter referred to as Optical Carrier (OC) links) used and the level of protection that is required.
One of the simplest types of line protection is a 1:1 linear protection scheme, as depicted in FIG. 1A, in which each working channel between two Network Elements (NEs) has a corresponding protection channel in parallel with it. These working and protection channels can be within a single OC link, though preferably they are within separate links as shown in FIG. 1A. The separation of the working and protection channels into two OC links allows an alternate route for the communications in the case that the OC link containing the working channel is disabled. The disadvantage of having separate links for the working and protection channels is the significant cost the additional OC links can add to the optical communication network.
In some configurations of line protection, a number of working channels between NEs share a single protection channel, an example of such a configuration being shown in FIG. 1B. These configurations are referred to as 1:n protection schemes where n corresponds to the number of working channels that depend upon a single protection channel. In the sample configuration of FIG. 1B, n is equal to three. The advantage of these 1:n protection architectures is the reduction in the number of OC links that are required for implementation. The key disadvantage is the reduced level of protection that is established. For instance, with such a protection architecture, the failure of two or more working channels that correspond to a single common protection channel is not protected against and will result in a non-correctable failure if no other protection scheme is in place.
Another protection technique that is commonly used within an optical communication network is a Bi-directional Line Switched Ring (BLSR). Within a network that is connected with a BLSR protection scheme, the NEs that typically comprise add/drop multiplexers are connected in a series configuration that loops into a circle as depicted in FIG. 2A. Essentially, when configured into a BLSR, communications from any one NE in the ring to any other NE in the ring can be directed in either the clockwise or counter-clockwise direction. This allows for a completely independent path of communications in the case that an OC link and/or NE is disabled. Even in the BLSR design, each working channel in each direction is typically protected with a protection channel such that in essence communications being transmitted via a working channel has a number of transmission options in cases that the working channel is disabled.
Typical BLSR designs come in two varieties, those that comprise two optical fibre cables that are referred to as 2F BLSRs and those that comprise four optical fibre cables that are referred to as 4F BLSRs. In a 2F BLSR, there is a single OC link between each pair of NEs for each direction through the ring, each OC link typically having its bandwidth divided equally between working and protection channels. In a 4F BLSR, there are two OC links between each pair of NEs for each direction through the ring, one for the working channel in the particular direction and the other for the protection channel.
Since the protection level and budget requirements are different from network to network, a BLSR design must be flexible and allow for numerous modifications. Different BLSR designs allow for modified balances between the number of OC links used and the protection level provided. For instance, there are instances in which BLSR networks allow for data traffic beyond the bandwidth of the working channel, hereinafter referred to as extra traffic, to be transmitted on the protection channel rather than requiring a working channel of larger bandwidth. Further, some BLSR networks allow for communications between particular NEs to be unprotected, this unprotected traffic being transmitted within the working and/or protection channels but with lower priority such that, if the bandwidth used is required for other purposes or a failure occurs in the particular channel being used for the unprotected traffic, the transmission of the unprotected traffic can be discontinued without serious problems. Yet further, similar to linear line protection, some or all of the connections between NEs of a BLSR could be implemented with a 1:n protection architecture in order to reduce the number of protection OC links that are required.
Although these modifications allow for an adjustable configuration for the BLSR architecture, it is recognized that these modifications also add to the overall complexity of the optical communication network and therefore the difficulty to manage the network. This complexity is especially prevalent when considering combinations of more than one of the above modifications within a single BLSR. There are further difficulties with a BLSR architecture even when no modifications from the standard design are required, one of which is now described with reference to FIGS. 2A and 2B.
FIG. 2A illustrates a situation in which data traffic (DATA1) is being transmitted within a BLSR from a first NE 50, via second and third NEs 52, 54, to a fourth NE 56. FIG. 2B illustrates the situation that occurs within the typical BLSR of FIG. 2A in the case in which a failure occurs within the OC link (containing both the working and protection channels) that connects the second and third NEs 52, 54. As depicted in FIG. 2B, in the case of the failure between the second and third NEs 52, 54, the data traffic (DATA1) being transmitted from the first NE 50 to the fourth NE 56 is re-routed around the failure. In a typical BLSR architecture currently used, this re-routing is done by sending the data traffic (DATA1) from the first NE 50 to the second NE 52, subsequently from the second NE 52, via first, fifth, sixth and fourth NEs 50, 58, 60, 56, to the third NE 54, and finally from the third NE 54 to the fourth NE 56. Although this re-routing allows the maintaining of communications between the first and fourth NEs 50, 56, the result of the line protection switching is a significantly inefficient use of the OC links between the first and second NEs 50, 52 and between the third and fourth NEs 54, 56. In both cases, the data traffic (DATA1) is double backing on its own path that decreases the available bandwidth within the effected OC links and further increases the time of transmission of the data traffic (DATA1) unnecessarily. This problem is caused by the fact that, in typical BLSR designs, when an OC link fails, all data traffic traversing the OC link is sent along the protection path corresponding to the failed OC link without consideration of the actual paths that the particular data traffic are traversing within the network.
There is a technique, referred commonly as transoceanic switching, that is occasionally utilized to remove the inefficiency described above while not moving away from a BLSR architecture. Transoceanic switching allows for the NEs at the start and end points for each data traffic path to be considered when re-routing data traffic after a working channel failure. This consideration essentially makes the line protection switching scheme of the BLSR into a combination between line protection and path protection architectures, reducing the inefficiencies associated with line protection while maintaining the standard BLSR framework. The problem with transoceanic switching, similar to the other possible modifications for a BLSR design, is the complexity that results from its implementation and the resulting difficulty in managing the overall network.
One technique that has been tried in order to remove the problems of BLSR designs and their numerous modifications that are commonly required is to move to a mesh protection design as illustrated in FIG. 3. In a full mesh design, each NE within a network is coupled to every other NE while in partial mesh designs, less OC links are utilized. The concept behind a mesh design is to establish working paths for all of the data traffic within the network while having a path protection strategy in place for any single failure. Within a well-known mesh design, when a failure occurs within a working path that has been established between two NEs, the network manager determines a new working path for the data traffic based upon the available bandwidth of the remaining OC links in the network. Well-known mesh techniques have an advantage in terms of minimizing the requirements for dedicated protection path bandwidth since the optical bandwidth used for protection is only assigned to the protection during a failure situation, hence reducing the cost of additional optical fibre cables.
One key problem with these well-known mesh designs is the amount of time that is required to locate and establish a new working path after a failure occurs. The time it takes to re-establish communications after failure is critical since the time period during switching should be small enough so as to be unnoticeable to the devices or people using the data traffic. In fact, the speed of protection switching is one of the primary advantages of the BLSR design described above, thus making the BLSR design extremely popular despite its problems with complexity when flexibility is added.
Hence, a new technique for protection switching within an optical communication network is required. Preferably, this new protection switching technique would have flexibility to adhere to specific customer requirements and would have somewhat comparable switching speeds to a standard BLSR design.